[HLSL] Mark GroupMemoryBarrierWithGroupSync as convergent (#160175)

`GroupMemoryBarrierWithGroupSync` is required to be marked as convergent
so that it can't generate duplicate calls or be moved to identical
control flow.

Without it, we generate undefined behaviour during optimization. For
instance: https://godbolt.org/z/9j3vsq1h3.

Testing that the convergent attribute is added is sufficient. There
already exists testing,
[here](https://github.com/inbelic/llvm-project/blob/main/llvm/test/Transforms/SimplifyCFG/attr-convergent.ll),
that it will not be moved as described in the above link.

Author: inbelic

clang/lib/Headers/hlsl/hlsl_alias_intrinsics.h

@@ -2805,7 +2805,7 @@ float4 radians(float4);
 /// call.
 
 _HLSL_BUILTIN_ALIAS(__builtin_hlsl_group_memory_barrier_with_group_sync)
-void GroupMemoryBarrierWithGroupSync(void);
+__attribute__((convergent)) void GroupMemoryBarrierWithGroupSync(void);
 
 } // namespace hlsl
 #endif //_HLSL_HLSL_ALIAS_INTRINSICS_H_

clang/test/CodeGenHLSL/builtins/GroupMemoryBarrierWithGroupSync.hlsl

@@ -17,4 +17,4 @@ void test_GroupMemoryBarrierWithGroupSync() {
 
 // CHECK: declare void @llvm.[[TARGET]].group.memory.barrier.with.group.sync() #[[ATTRS:[0-9]+]]
 // CHECK-NOT: attributes #[[ATTRS]] = {{.+}}memory(none){{.+}}
-// CHECK: attributes #[[ATTRS]] = {
+// CHECK: attributes #[[ATTRS]] = {{.+}}convergent{{.+}}

llvm/include/llvm/IR/IntrinsicsDirectX.td

@@ -160,5 +160,6 @@ def int_dx_firstbituhigh : DefaultAttrsIntrinsic<[LLVMScalarOrSameVectorWidth<0,
 def int_dx_firstbitshigh : DefaultAttrsIntrinsic<[LLVMScalarOrSameVectorWidth<0, llvm_i32_ty>], [llvm_anyint_ty], [IntrNoMem]>;
 def int_dx_firstbitlow : DefaultAttrsIntrinsic<[LLVMScalarOrSameVectorWidth<0, llvm_i32_ty>], [llvm_anyint_ty], [IntrNoMem]>;
 
-def int_dx_group_memory_barrier_with_group_sync : DefaultAttrsIntrinsic<[], [], []>;
+def int_dx_group_memory_barrier_with_group_sync
+    : DefaultAttrsIntrinsic<[], [], [IntrConvergent]>;
 }

llvm/include/llvm/IR/IntrinsicsSPIRV.td

@@ -127,7 +127,8 @@ def int_spv_rsqrt : DefaultAttrsIntrinsic<[LLVMMatchType<0>], [llvm_anyfloat_ty]
       : DefaultAttrsIntrinsic<[llvm_i32_ty], [], [IntrConvergent]>;
   def int_spv_sign : DefaultAttrsIntrinsic<[LLVMScalarOrSameVectorWidth<0, llvm_i32_ty>], [llvm_any_ty], [IntrNoMem]>;
   def int_spv_radians : DefaultAttrsIntrinsic<[LLVMMatchType<0>], [llvm_anyfloat_ty], [IntrNoMem]>;
-  def int_spv_group_memory_barrier_with_group_sync : DefaultAttrsIntrinsic<[], [], []>;
+  def int_spv_group_memory_barrier_with_group_sync
+      : DefaultAttrsIntrinsic<[], [], [IntrConvergent]>;
   def int_spv_discard : DefaultAttrsIntrinsic<[], [], []>;
   def int_spv_uclamp : DefaultAttrsIntrinsic<[llvm_anyint_ty], [LLVMMatchType<0>, LLVMMatchType<0>, LLVMMatchType<0>], [IntrNoMem]>;
   def int_spv_sclamp : DefaultAttrsIntrinsic<[llvm_anyint_ty], [LLVMMatchType<0>, LLVMMatchType<0>, LLVMMatchType<0>], [IntrNoMem]>;